© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 61632
    Prev
    Next
    INDEX
    Explanations

    full description

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    èĤ¢
    -0.10
    è¶ĭåĬ¿
    -0.08
    辩æĬ¤
    -0.08
    -Ta
    -0.08
    çļĦè¶ĭåĬ¿
    -0.08
    æĸĩåĮĸ建设
    -0.08
    个ä½ĵ
    -0.08
    被迫
    -0.08
    橱
    -0.08
    åİĨ代
    -0.08
    POSITIVE LOGITS
    fox
    0.09
    æį»
    0.09
    ä¾µ
    0.09
    (KP
    0.09
    CAL
    0.08
    -component
    0.08
    çµ®
    0.08
    sink
    0.08
    åĿŀ
    0.08
     Battlefield
    0.08
    Activations Density 0.017%

    No Known Activations